Latent Space: The AI Engineer Podcast
01:51:03 · S2, Mihail Eric, Shawn Wang
[01:51:03] S2: I think at some level students should also stay grounded and like, there's a lot of stuff happening at the edge. There's also a lot of companies that are still like single call tsetsepticeye production, playing around, haven't tried it. There is a lot of that. So there's still time for what to work. I think your point on taste is good, have taste, try stuff like whether each company has their own version of a software factory, you're building tools to help companies do it, staying on top of it, actually using the tools playing with it, like, you'll you'll figure it out along the way, you know. [01:51:41] S1: Yeah. And I think the the 100% agree with all that. I I mean, first, to your point that are a lot of companies still in the prompt ChatGPT or some version of, you know, ChatGPT? Maybe they just got a cursor license and are just now getting people onboarded. That is way more true than we believe, certainly here in the Valley. That is... Can I can say 100% confidence? But, yes, I think for the students being exposed to this, it's just at a certain point, this... The fact that they know this and are on the edge is what will eventually allow them to also be some of the more interesting and more valuable contributors when they go to maybe places where... That maybe aren't up to snuff in terms of being on the frontier. [01:52:16] S0: Yeah. One of the... This this kind of progression type talk, still very influential, about a year old Steve Yegge Mhmm. Talking about the the levels of psychosis. He's all the way through level 10 now [01:52:27] S2: where he's
01:52:27 · S2, Shawn Wang, Mihail Eric
[01:52:27] S2: where he's [01:52:27] S0: he's he's given up and he's moved to a a farm. But... Yeah. Like, you want you want... Sorry. I'm I'm making fun of him. He built Gastown. [01:52:36] S1: Yeah. I remember. [01:52:36] S0: Yeah. All these things, moved to Gas City. Now he's burned down the city and said, I'm building my game. I'm... No more Gastown. [01:52:43] S1: I thought I thought that... I mean, I I remember talking to him at at the Engineer Fair earlier this year. And wasn't he... His whole, like, thing, like, the only useful thing that was built at Gastown was Gastown. Yeah. [01:52:53] S0: Yeah. Okay. Right. With with that, that's a horrible failure. Horrible. [01:52:56] S1: Totally. Totally. [01:52:58] S0: And and so, like, people should be aware that, like, psychosis is real. And you... If you're not holding yourself accountable to delivering real products, you're just screwing along with LMs. [01:53:08] S1: Yeah. This is like a... Maybe, I mean, relevant to... But I wonder how much of... I mean, played a little bit with Gastown. Very involved system. Maybe a little ahead of time. But I'd also wonder if they were, like, bottlenecked a little bit by the by the marketing, and that it was a very difficult thing to grok from a terminology perspective. There's just a lot going on. [01:53:24] S0: It made sense to him. [01:53:25] S2: I think the Gastown marketing also is one of the things that propelled it. Right? Like, what was right before Gastown? There were like... [01:53:35] S1: Like, yeah. His... Yeah. [01:53:36] S2: Beats. Yeah. Did. Unless you're
01:53:36 · S2, Shawn Wang
[01:53:36] S2: Beats. Yeah. Did. Unless you're [01:53:37] S0: talking with someone else. No. [01:53:38] S2: No. There were like... There was a whole open claw. There was something else and there was Gastown. And the whole manifesto also really helped spur it into the conceptual. The people that would never use Gastown but learned about Gastown because you can read the manifesto. Like, there... There's a marketing angle to it.
01:53:55 · Shawn Wang
[01:53:55] S0: Yeah. I mean... So so I... But I do... So Steve is not necessarily someone to take literally, but direct... Like like, you know, directionally correct. You know? Yeah. And and so I think he's directionally correct on this one. And so if you're at the IDE stage, he's trying to move into a power user stage. Yep. If you're a single agent, you're probably trying to move to multi agent and blah blah blah. And you probably should try to build your own orchestrator. Right? That's it's all that. Arguably, this is kind of out of date now. There is more to The Software Factory beyond this, but this article predates The Software Factory. I do think one of those things where I'm open to... As an employer, I have this problem in my company today, which is I do not want to hire anyone who hasn't been through psychosis and come out the other end. Because I am currently paying for somebody to go through psychosis. And it's very expensive from his salary, very expensive from tokens, I'm getting nothing out of it and he doesn't get it. Like my feedback doesn't get through. And so like, it's almost like I would encourage your students to have psychosis and then come out and go like, okay, what was wrong with me? Because like, that's what Steve just went through, like a very public two year journey of this.
01:55:02 · S2, Mihail Eric, Shawn Wang
[01:55:02] S2: Yeah. It's weird on the student side. There's like a lot of industry, you think you're being productive, you're not, it's probably the student version, right? You're notion charting and making all these optimizations, personal assets like, I went through the psychosis, but nothing has changed. Right? [01:55:18] S1: Right. Right. [01:55:18] S2: So go through it. Go through it. [01:55:19] S1: Right. Yeah. I haven't seen this. I've seen a version of this from Boris, actually. Obviously, more geared toward cloud use cases, but I remember he made, like, a cloud artifact, which showed, I think more of it at the team level, but I think he had four stages, not eight stages. But it kind of rough... Roughly followed the same progression of, like, no adoption to... Through to, [01:55:37] S0: you know, I guess Yeah. Only four is good. This is a... There's too many levels. Yeah. Well, see... We'll see if Vivo can pull it up. But other than that, you know, zooming out. Have we, you know, discard... Discussed all the things? That's one. Oh, that's one? Okay. Yeah. Here we go. It is an artifact. Okay. So it's gated, assisted, parallel. What's the last one? Supervised autonomy and AI native. Okay. [01:55:57] S1: So it's zero index. So it's five levels. Right? [01:55:59] S0: Alright. Insert zero. Yeah. Wow. Actually, it's pretty good. I need to redo this in more more depth. [01:56:07] S1: Can you say native?
01:56:07 · Mihail Eric, Shawn Wang, S2
[01:56:07] S1: Can you say native? [01:56:08] S0: Yeah. Yeah. I I would give a shout out to dynamic workflows. This is one of those things that is a sub agent pattern Yeah. That the opening guys are also exploring that is clearly fitting. And and so it's basically you're writing a harness on demand for the thing that you're trying to do. Yep. And so this is maybe... Like, a lot of sub agent conversation is just you're copying the the parent agent. Yep. But you actually should be writing way more code and then, like, running that as a as a as a sub agent, and that is what the Anthropic people have chosen to call dynamic workflows. Okay. But... You know? [01:56:45] S1: Yeah. We... We've built a version as well. Yeah. It's it's a good pattern. [01:56:48] S0: For Monaco? Mhmm. [01:56:49] S1: Yeah. [01:56:49] S0: What does it do? [01:56:50] S1: Yeah. I mean, on the fly, it's really, harness sort of use cases for GCAM. [01:56:54] S2: Yeah. Oh, that's great. Yeah. I will note that when Boris wrote this, it's for team adoption for people that are listening. And also he says, Anthropic is on stage three. We're still trying to get to stage four. Wow. He's like, I am at stage four. I'm the guy. But it's not as simple as Steve's. Like, Steve is a very good personal do this, do this, do this. This is a much more... If you're running a team and you wanna AI pill your team and not just be a user of one. Anthropic is still at stage three. Don't don't be like... Yeah. Yeah. Okay.